Metformin alleviates lung-endothelial hyperpermeability by regulating cofilin-1/PP2AC pathway

Abstract

Background: Microvascular endothelial hyperpermeability is an earliest pathological hallmark in Acute Lung Injury (ALI), which progressively leads to Acute Respiratory Distress Syndrome (ARDS). Recently, vascular protective and anti-inflammatory effect of metformin, irrespective of glycemic control, has garnered significant interest. However, the underlying molecular mechanism(s) of metformin's barrier protective benefits in lung-endothelial cells (ECs) has not been clearly elucidated. Many vascular permeability-increasing agents weakened adherens junctions (AJ) integrity by inducing the reorganization of the actin cytoskeleton and stress fibers formation. Here, we hypothesized that metformin abrogated endothelial hyperpermeability and strengthen AJ integrity via inhibiting stress fibers formation through cofilin-1-PP2AC pathway. Methods: We pretreated human lung microvascular ECs (human-lung-ECs) with metformin and then challenged with thrombin. To investigate the vascular protective effects of metformin, we studied changes in ECs barrier function using electric cell-substrate impedance sensing, levels of actin stress fibers formation and inflammatory cytokines IL-1β and IL-6 expression. To explore the downstream mechanism, we studied the Ser³-phosphorylation-cofilin-1 levels in scramble and PP2AC-siRNA depleted ECs in response to thrombin with and without metformin pretreatment. Results: In-vitro analyses showed that metformin pretreatment attenuated thrombin-induced hyperpermeability, stress fibers formation, and the levels of inflammatory cytokines IL-6 and IL-β in human-lung-ECs. We found that metformin mitigated Ser³-phosphorylation mediated inhibition of cofilin-1 in response to thrombin. Furthermore, genetic deletion of PP2AC subunit significantly inhibited metformin efficacy to mitigate thrombin-induced Ser³-phosphorylation cofilin-1, AJ disruption and stress fibers formation. We further demonstrated that metformin increases PP2AC activity by upregulating PP2AC-Leu³⁰⁹ methylation in human-lung-ECs. We also found that the ectopic expression of PP2AC dampened thrombin-induced Ser³-phosphorylation-mediated inhibition of cofilin-1, stress fibers formation and endothelial hyperpermeability. Conclusion: Together, these data reveal the unprecedented endothelial cofilin-1/PP2AC signaling axis downstream of metformin in protecting against lung vascular endothelial injury and inflammation. Therefore, pharmacologically enhancing endothelial PP2AC activity may lead to the development of novel therapeutic approaches for prevention of deleterious effects of ALI on vascular ECs.

Keywords: PP2A; acute lung injury; cofilin-1; metformin; vascular endothelial cells.

FIGURE 1

Metformin attenuates thrombin-induced endothelial hyperpermeability and stress fibers formation (A) Quiescent human lung-ECs were left untreated or treated with 10 mM of metformin for indicated time points. Equal amount of protein lysates from each condition were analyzed by Western blotting for phospho-AMPK-Thr¹⁷² and actin. Data was performed in three independent experiments (B) Quiescent human lung-ECs were analyzed for electrical cell impedance (represented as fold change index). ECs were treated with PBS, thrombin (25 nM) or pretreated with metformin (5 and 10 mM for 2 h) followed by thrombin challenge. Data analysis was performed in triplicate, and data are presented as mean ± SD; ^∗∗ p < 0.01 vs. thrombin-treated group (C) Quiescent human lung-ECs were left untreated or treated with thrombin (25 nM) for indicated time points in the absence and presence of metformin (10 mM, 2 h). Representative immunofluorescence images of F-actin labelled with phalloidin-FITC (Scale bar: 20 μm). Data was performed in three independent experiments (D) qRT-PCR analysis of proinflammatory mediators. ECs were left untreated or treated with thrombin (25 nM) for indicated time points in the absence and presence of metformin (10 mM, 2 h). Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗ p < 0.05, ^∗∗ p < 0.01 vs. thrombin-treated corresponding time group.

FIGURE 2

Metformin attenuates thrombin-induced cofilin-1 phosphorylation. Quiescent human lung-ECs were left untreated or treated with thrombin (25 nM) for indicated time points in the absence and presence of metformin (10 mM, 2 h) and equal amount of protein lysates from each condition were analyzed by Western blotting for phospho-cofilin-1-Ser³ and actin. The bar graphs represent quantitative analysis. Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗∗ p < 0.001 vs. thrombin-treated group.

FIGURE 3

PP2AC silencing impairs metformin efficacy for cofilin-1 dephosphorylation to protect endothelial barrier function (A) Human lung-ECs were transfected with scramble-siRNA (sc-siRNA) or PP2AC-siRNA then left untreated or treated with thrombin (25 nM, 15 min) in the absence and presence of metformin (10 mM, 2 h). Equal amounts of proteins from each condition were analyzed by Western blotting for phospho-cofilin-1-Ser³ and actin. The bar graphs represent quantitative analysis. Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗ p < 0.001, ^∗ p < 0.05 (B) Human lung-ECs were transfected with sc-siRNA or PP2AC-siRNA then left untreated or treated with thrombin (25 nM) in the absence and presence of metformin (10 mM, 2 h). Electrical cell impedance was analyzed (represented as fold change index). Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗ p < 0.001 vs. Scr-siRNA + thrombin treated group, ^∗ p < 0.05 vs. Scr-siRNA + thrombin, ##p < 0.001 vs. Scr-siRNA + met + thrombin treated group (C) Human lung-ECs were transfected with sc-siRNA or PP2AC-siRNA then left untreated or treated with thrombin (25 nM, 30 min) in the absence and presence of metformin (10 mM, 2 h). Representative immunofluorescence images of F-actin labelled with phalloidin-FITC (Scale bar: 20 μm). Data was performed in three independent experiments.

FIGURE 4

Metformin induced PP2AC phosphatase activity (A) Phosphatase activity was determined as described in the experimental procedure. Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗∗ p < 0.001 vs. untreated group. Quiescent human lung-ECs were treated with 10 mM of metformin for the indicated time points. Equal amounts of proteins from each condition were analyzed by Western blotting for (B) phospho-PP2AC-Y³⁰⁷ and PP2AC (C) methyl-PP2AC-Leu³⁰⁹ and PP2AC. The bar graphs represent quantitative analysis. Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗∗ p < 0.001 vs. 0 h treated group.

FIGURE 5

Overexpression of PP2AC mitigated thrombin-induced hyperpermeability and cofilin-1 phosphorylation. Human lung-ECs were transfected with PP2AC cDNA for 48 h and then (A) left untreated or treated with thrombin for indicated time points. Equal amount of proteins from each condition were analyzed by Western blotting for phospho-cofilin-1-Ser³, GFP and actin. The bar graphs represent quantitative analysis Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗∗ p < 0.001 vs. non-transfected thrombin-treated corresponding time point. Human lung-ECs were transfected with PP2AC cDNA for 48 h and then (B) left untreated or treated with thrombin (25 nM) and electrical cell impedance was analyzed (represented as fold change index). Data analysis was performed from three independent experiments, and are presented as mean ± SD; ^∗∗ p < 0.01 vs. non-transfected thrombin-treated group (C) Left untreated or treated with thrombin (25 nM, 30 min) in the absence and presence of metformin (10 mM, 2 h). Representative immunofluorescence images of F-actin labelled with phalloidin-FITC (Scale bar: 20 μm). Data was performed in three independent experiments.